Current flow into a charging battery?

[stargazer193857]

I know how to calculate current flow across a resister from the voltage, but what about current flow into a battery? Do I need to put a resister in the system so that I know how much current is flowing, or does that drop the voltage to the battery too much?

I'm guessing the battery has an effective resistance to it, and as long as the voltage going to it is within specs, the current will be safe as well, and decrease as the battery fills. I'm guessing no resister is needed, and would actually reduce the current flow well below that predicted by the voltage and resistor alone. But I just want to make sure.

 

[mfb]

I'm guessing the battery has an effective resistance to it

Right, but I would not rely on that even if your voltage cannot exceed the design voltage - the battery could heat up if it gets too much current.

 

[stargazer193857]

Well, before I do any experiments, I would put a fuse in series with the battery, one that melts just above the goal current.

The simple charger circuits I looked at don't have resisters in series with the battery. My calculations say that if the resister limits the current to the goal flow, it zeros the voltage difference between the charger and the battery. So I need to know the actual resistance of the battery in order to calculate the flow. We know that a voltage difference must be given to the battery for it to charge. That voltage difference would cause unlimited flow if the battery has zero resistance. So I think that means the battery does have significant resistance, and as long as the charger is in the published voltage range, a resister might not be needed.

I would like some people in the electrical engineering forum to give their input on this.

 

[mfb]

Commercial chargers usually monitor the current flow and adjust their power source accordingly.
If you apply exactly the design voltage, then you might get no charging effect at all, or it might take a really long time.

Electrical engineering sounds good, I moved the thread.

 

[sophiecentaur]

Well, before I do any experiments, I would put a fuse in series with the battery, one that melts just above the goal current.

The simple charger circuits I looked at don't have resisters in series with the battery. My calculations say that if the resister limits the current to the goal flow, it zeros the voltage difference between the charger and the battery. So I need to know the actual resistance of the battery in order to calculate the flow. We know that a voltage difference must be given to the battery for it to charge. That voltage difference would cause unlimited flow if the battery has zero resistance. So I think that means the battery does have significant resistance, and as long as the charger is in the published voltage range, a resister might not be needed.

I would like some people in the electrical engineering forum to give their input on this.

The resistance will probably be built into the transformer windings with thinner wire. Cheaper to build a rubbish transformer and save on an extra resistive component. A 'good battery' can melt thin wires and nails so you can't rely on the battery internal resistance.

 

[stargazer193857]

I need 20 or 22g wire to be completely safe at 3A, assuming I want to the transformer to not melt. I don't think I'll get all the windings of a normal transformer. Do I really need a ton of windings, or just the right ratio on each side? Will more windings be more efficient? I'm not making a super strong electromagnet. I'm just trying to change the voltage. I wonder if 120 wraps and 40 wraps is good enough. Or 60 and 20. Winding takes time.

 

[meBigGuy]

You need to design the charger to be compatible with the battery technology. If you are charging LI-ion batteries, you are in danger of causing fire or explosions if you do not use a LI-ion battery charge controller, or fully follow the 3 phase LI-ion charge profiles and precision measurement requirements.

I recommend you search for the proper charge profile for the battery technology you are using. There are multiple battery manufacturer pages for any technology you want to use, and many IC application notes for charger controllers. Once you have read a bit, come back and ask questions, and be sure to describe more about the battery and the system architecture you are aiming for.

 

[stargazer193857]

Well, I found out from an unhappy review that commercial chargers for 36v batteries charge at 42V. This is true for lead acid and Lithium ion. The difference is it is safe to leave the lead acid on all night, and needed. The reviewer left his lithium ion on too long and it caught on fire. He thought the voltage was too high. It was not too high. But it was not a smart charger, and he should have taken it off sooner. Smart chargers monitor the current flow and cut it off when it reaches a trickle.

 

[stargazer193857]

You need to design the charger to be compatible with the battery technology. If you are charging LI-ion batteries, you are in danger of causing fire or explosions if you do not use a LI-ion battery charge controller, or fully follow the 3 phase LI-ion charge profiles and precision measurement requirements.

I recommend you search for the proper charge profile for the battery technology you are using. There are multiple battery manufacturer pages for any technology you want to use, and many IC application notes for charger controllers. Once you have read a bit, come back and ask questions, and be sure to describe more about the battery and the system architecture you are aiming for.

Yes. For a 36v Li-NMC battery, 42v is used to charge it. However, the batter is supposed to be taken off when it reaches 40.5v. 42v actually is not enough to catch it on fire, though a millivolt more might. Charging it to 42v would drastically reduce the number of remaining cycles, and not extend the amp hours for the cycle by much. A commercial charger rated for lithium ion batteries monitors the current flow and cuts it off once it slows to the point associated with 40.5v. I could get close enough just by leaving it on the charger for only 2-3 hours.

I'm already buying copper wire, solder wire, and a soldering gun, and electrical tape to connect 27 cells in series and parallel, so I might as well have fun and make my own charger too. The extra parts for the charger are only $5 more.

I'll just make sure not to ask the motor for full power after I've been riding 30 minutes, that way it does not have to draw extra current to make up for 10% less voltage.

 

[stargazer193857]

I will post a picture of my circuit drawing for the charger before I build it, and I'll post a picture of the completed charger before I use it. I was going to ask for permission to use one of my university shops. But since the soldering iron is so cheap, I think I'll do it at home. I'll post pictures of intermediate steps, and time myself.

The Li-NMC battery cells are $2.50 for 2500mAH. 9 in series fully charged makes 36v. 3 of these in parallel makes 7.5AH. 1C is very doable. 7.5Ax36v = 270W, which is more than the 250 motor, so I should be good. 27 cells x $2.50 plus $20 shipping was about $80, and much less than the $500 for a commercial pack of similar size.

I'd make them for resale, but I know most people don't want to pay much. Ebay or Amazon would give me easy advertising, though. I'm sure there are regulations from certifying boards that would prevent me from doing this. I would hate for someone to leave their battery charging over night, let it catch on fire, and then try to sue me for it. Depending on what state they live it they might get away with it.

 

[meBigGuy]

You need charge balancing circuits for charging multiple li-ion batteries. I think you need to do a lot more research. At least, try to determine what charge balancing is all about.

 

[stargazer193857]

I think since my three series are in parallel, they will be charged the same. No balancing needed.

I think I figured out the answer to my question. A battery is like a diode. It has very little resistance one direction and infinite the other direction. Any energy that goes in the charging direction goes to charging the battery, or making local gas bubbles if too much at once.

A diode is like a battery with lots of internal self discharge. It does not build up energy.

The fact that a battery can melt a nail during rapid discharge and has an internal resistance of only 30 miliOhms does not mean it will allow a ton of current the other direction when faced with a voltage. If you put 12.8 volts on an uncharged lead acid battery, no current will flow, and no charging occurs. You must first pass the threshold. It is like pushing a sled. You first must overcome static friction. My batteries will do just fine with a 42 volt charger. I have decide to buy an $18 multimeter so I know how full they are, and can test my charger too. It is also nice having the meter.

I will buy all the parts once I know what kind of connector the motor takes, so I can order everything at once.

 

[meBigGuy]

You are very very wrong about quite a few things, and will destroy the batteries, possibly injure yourself. Charging 9 lithium batteries in series requires charge balancing.
http://www.digikey.com/videos/en/v/Lithium-Ion-Battery-Cell-Balancing/1646398576001
http://en.wikipedia.org/wiki/Battery_balancing

Even paralleling single cells requires matching for decent battery life
http://web.mit.edu/bazant/www/papers/pdf/Gogoana_2013_J_Power_Sources.pdf

 

[sophiecentaur]

I think since my three series are in parallel, they will be charged the same. No balancing needed.

I think I figured out the answer to my question. A battery is like a diode. It has very little resistance one direction and infinite the other direction. Any energy that goes in the charging direction goes to charging the battery, or making local gas bubbles if too much at once.

A diode is like a battery with lots of internal self discharge. It does not build up energy.

The fact that a battery can melt a nail during rapid discharge and has an internal resistance of only 30 miliOhms does not mean it will allow a ton of current the other direction when faced with a voltage. If you put 12.8 volts on an uncharged lead acid battery, no current will flow, and no charging occurs. You must first pass the threshold. It is like pushing a sled. You first must overcome static friction. My batteries will do just fine with a 42 volt charger. I have decide to buy an $18 multimeter so I know how full they are, and can test my charger too. It is also nice having the meter.

I will buy all the parts once I know what kind of connector the motor takes, so I can order everything at once.

If that were true then how could you ever recharge a battery? Have you ever tried to see what charging current a rechargeable cell will take? (Don't try)
I suggest you buy a "sled" and play with that. It will be safer than some of the things you are proposing to do with your batteries. :D

Very good idea to buy a meter, though. :)

 

[meBigGuy]

PLEASE DO NOT TRY WHAT YOU ARE TALKING ABOUT DOING. YOU WILL CAUSE A FIRE. YOUR MODEL IS TOTALLY WRONG.
http://batteryuniversity.com/learn/article/lithium_ion_safety_concerns
http://cds.linear.com/docs/en/design-note/dn188f.pdf

Let's talk about how to charge 1 Li-ion cell, and how the cell responds to applied voltage.

There are three phases to charging a li-ion battery. The correct voltages and currents may vary based on the battery pack specifications and manufacturer. Not following the protocol can damage the cell or cause a fire.

1. Precharge. If the cell voltage is below 3.0V, you must apply a small current (about 10% of the specified charge current for the battery) until the cell reaches 3 volts. If your cells contain a safety pack (I THINK most do), they will probably not be below the voltage required to begin the constant current phase.

2. Constant current. During this phase you apply a constant current. The amount varies based on the manufacturer specifications, but is typically 1C. So, for a 1200ma battery that would be 1.2A. This requires a constant current power supply. If you start with the battery at 3.0V, this will occur when you apply slightly above 3.0V to the battery. As it charges, the required voltage changes. For example, when the battery reaches 3.7V, you will need to apply slightly more than 3.7V to maintain a 1C rate. You continue this phase until the battery voltage reaches 4.2V +- 50mv. As I said, the applied voltage required to supply a 1C current changes as the battery charges. Supplying too much current can cause a fire.

3. Constant voltage. During this phase you must maintain the voltage across the battery to 4.2V +- 50mv (depends on the manufacturer) by reducing the charge current. When the charge current reaches 0.1C you can terminate charging. Skipping this phase will result in a significantly undercharged battery.

Connecting batteries in series complicates this because not all cells will be at the exact same voltage. So charge balancing is required.

Connecting batteries in parallel also causes issues because during the constant current phase you can't control how much current goes into each battery. So you must limit to 1C and charge slowly.

Some of this can be avoided by charging at a very low contant current.

Let's say you have 9 1200mah cells discharged to 3.0V each. That requires somthing a little above 27V to charge at 1C. Whn they reach 4.0V per cell, you need slightly above 36V to supply 1C. Once the voltage reaches 4.2V per cell you are in contant voltage mode.

I think you are in way over your head.

 

[stargazer193857]

Thank you for those informative links. I think I'll have two wires attached to each terminal: one to get it in series, and another for connecting all positives to each other and all negatives to each other before and after each charging, when not on the bike, so they can even each other out. I'm not yet done reading all the links, but I could have a switch and more than one transformer so that I can vary the voltage throughout the charge cycle, checking them periodically with my multimeter. Also, I plan to build a bigger battery that won't come close to discharging, so I stay far away from 3.0 volts. If I only operate between 4.0v and 3.6v, I think I'll be fine.

 

[stargazer193857]

One fact in one article:
Lithium ion batteries have caused phones and laptops to explode if made with non-name brand batteries that don't have the internal protections. In many countries it is illegal to sell sells without a protection circuit, or to hobbyists. To make and sell battery packs commercially, one must be licensed to do so with lithium ion packs. Hobbyists better stick to their own projects or risk legal action. And if their pack catches on fire in public, and they are the ones who built it, that is another law suit they can lose.

 

[stargazer193857]

Another safety test is the feel test. If the cell feels hot, turn off the charger. Unfortunately, packs often don't allow a feel. Many commercial battery packs in case each cell in a heat conductive material that dissipates the heat of anyone cell. There are plenty for sale though with just duck tape.

 

[berkeman]

Thread closed for Moderation...

@stargazer193857 -- what you are trying to do is extremely dangerous for a person with so little electronics experience and training. Keep in mind you are endangering not only yourself, but others around you. If you start a fire and cause damage to property, you will be liable for payment to fix the damages. If you hurt somebody with your experiments, you could go to jail. Not good.